Modern data centers aim to offer very high throughput and ultra-low latency to meet the demands of applications such as online intensive services. Traditional TCP/IP stacks cannot meet these requirements due to their high CPU overhead and high-latency. Remote Direct Memory Access (RDMA) is an approach that can be designed to meet this demand. The mainstream transport protocol of RDMA over Ethernet is RoCE (RDMA over Converged Ethernet), which relies on Priority Flow Control (PFC) within the network to enable a lossless network. However, PFC is a coarse-grained protocol which can lead to problems such as congestion spreading, head-of-the-line blocking. A congestion control protocol that can alleviate these problems of PFC is needed. We propose a protocol, called P4QCN for this purpose. P4QCN is a congestion control scheme for RoCE and it is an improved Quantized Congestion Notification (QCN) design based on P4, which is a flow-level, rate-based congestion control mechanism. P4QCN extends the QCN protocol to make it compatible with IP-routed networks based on a framework of P4 and adopts a two-point algorithm architecture which is more effective than the three-point architecture used in QCN and Data Center QCN(DCQCN). Experiments show that our proposed P4QCN algorithm achieves the expected performance in terms of latency and throughput.

No abstract

Object Detection has become an important feature in field of Computer Science. Benefits of object detection are varied and are not restricted to any specific area. Instead, object detection technique is growing rapidly and used widely in Information Industry. This paper intends to address one such possibility with the help of Haar-cascade classifier. The focus of the paper will be on the case study to develop a Smart and Dynamic Traffic Management System which is based on vehicle detection mechanism.

Big data" has become a hot spot of the times, and the processing of big data is inseparable from Data Center Networks (DCN). As the infrastructure of the information era, data center networks provide a variety of network services and become key support technologies for future Internet/Cloud computing services and applications. In the process of soliciting data and using data, the performance of the data center network is very important. In the data center network communication process will produce a large number of TCP flows, if you can speed up the transmission of TCP flows, you can greatly reduce the response time, bring economic benefits; the other hand, if you can't properly control the data center network appears With a large number of TCP streams, there will be network congestion. This paper analyzes the problem of TCP fairness in data center network theoretically, designs the network topology according to the characteristics of the data center network transmission process, uses NS2 simulation platform to carry out experiments, and simulates and analyzes TCP fairness according to the principle of maximum and minimum fairness. The relationship between sex and bottleneck bandwidth, network parameters such as RTT, and the number of flows and other factors, and finally gives relevant conclusions.

No abstract